BIAS CALCULATIONS FOR ADAPTIVE URN DESIGNS

A clinical trial model is considered in which two treatments with immediate binary responses are to be compared. An adaptive urn design is used to assign patients to the treatments. The bias and variance of the maximum likelihood estimators of the probabilities of success are derived by differentiating the fundamental identity of sequential analysis. By embedding the design in a continuous-time process, probability generating functions are then calculated to obtain approximations for the bias and variance. Simulation is used to assess the accuracy of the approximations. It is shown that the bias cannot be ignored, and that the adaptive rules which are subcritical in nature have the most mathematically tractable bias and are the least variable. Methods for correcting for the bias are also addressed.

[1]  J. Fleiss.  Statistical methods for rates and proportions , 1974 .

[2]  A. Ivanova,et al.  Adaptive Designs for Clinical Trials with Highly Successful Treatments , 2001 .

[3]  M. Woodroofe.  On stopping times and stochastic monotonicity , 1990 .

[4]  Quentin F. Stout,et al.  Exact computational analyses for adaptive designs , 1995 .

[5]  William F. Rosenberger,et al.  New directions in adaptive designs , 1996 .

[6]  J. Lindsey.  Stopping rules and the likelihood function , 1997 .

[7]  H. Robbins.  Some aspects of the sequential design of experiments , 1952 .

[8]  S. D. Durham,et al.  A sequential design for maximizing the probability of a favourable response , 1998 .

[9]  Douglas E. Faries,et al.  A Case Study of an Adaptive Clinical Trial in the Treatment of Out-Patients with Depressive Disorder , 1994 .

[10]  William F. Rosenberger,et al.  Asymptotic normality of maximum likelihood estimators from multiparameter response-driven designs , 1997 .

[11]  B. Everitt,et al.  Statistical methods for rates and proportions , 1973 .

[12]  L. J. Wei,et al.  Statistical Inference with Data-Dependent Treatment Allocation Rules , 1990 .

[13]  Connie Page,et al.  Estimation after adaptive allocation , 2000 .

[14]  J. Haldane,et al.  On a method of estimating frequencies. , 1945, Biometrika.

[15]  Quentin F. Stout,et al.  Flexible Algorithms for Creating and Analyzing Adaptive Sampling Procedures , 1998 .

[16]  R. G. Cornell,et al.  Extracorporeal circulation in neonatal respiratory failure: a prospective randomized study. , 1985, Pediatrics.

[17]  M. Woodroofe.  Nonlinear Renewal Theory in Sequential Analysis , 1987 .

[18]  D S Coad,et al.  A comparison of the randomized play-the-winner rule and the triangular test for clinical trials with binary responses. , 1999, Statistics in medicine.

[19]  Nancy Flournoy,et al.  A Birth and Death Urn for Ternary Outcomes: Stochastic Processes Applied to Urn Models , 2000 .

[20]  L. J. Wei,et al.  The Randomized Play-the-Winner Rule in Medical Trials , 1978 .

[21]  Z. Govindarajulu,et al.  Corrected confidence intervals following a sequential adaptive clinical trial with binary responses , 2000 .

[22]  N. Cressie,et al.  The Moment-Generating Function and Negative Integer Moments , 1981 .

[23]  David R. Cox.  The analysis of binary data , 1970 .

[24]  Nigel Stallard,et al.  Exact group‐sequential designs for clinical trials with randomized play‐the‐winner allocation , 2002, Statistics in medicine.

[25]  Lee-Jen Wei,et al.  Play the winner for phase II/III clinical trials. , 1996, Statistics in medicine.

[26]  M. Woodroofe,et al.  Approximate bias calculations for sequentially designed experiments , 1998 .

[27]  John Bather,et al.  Sequential procedures for comparing several medical treatments , 1992 .

[28]  William F. Rosenberger,et al.  Variance in randomized play-the-winner clinical trials , 1997 .

[29]  M. Zelen,et al.  Play the Winner Rule and the Controlled Clinical Trial , 1969 .

[30]  Connie Page,et al.  Variability in Adaptive Designs for Estimation of Success Probabilities , 1998 .

[31]  W F Rosenberger,et al.  Bootstrap methods for adaptive designs. , 1999, Statistics in medicine.

[32]  W F Rosenberger,et al.  A COMPARISON OF URN DESIGNS FOR RANDOMIZED CLINICAL TRIALS OF K > 2 TREATMENTS , 2000, Journal of biopharmaceutical statistics.

[33]  D. S. Coad.  Data-dependent allocation rules a comparative study of some for bernoulli data , 1992 .